The present invention relates to an image processing apparatus and an image forming apparatus, which function to store image data obtained by reading an image, into a memory, and which forming an image on a paper sheet by subjecting the image data to image processing.
There has been a conventional image forming apparatus in which scanning using a laser beam is carried out by a polygon motor and the laser beam is irradiated onto a photosensitive material thereby forming an electrostatic latent image, which is visualized by a developing step and is transferred as an image onto a recording medium by a fixing means. This technique is frequently used in laser printers, plain paper copiers (PPC) in recent years.
It is necessary that pixels have respectively desired sizes in case where an electrostatic latent image is formed by switching on or off the laser for every pixel and is visualized by fine toner.
For example, to express an image at 400 dpi, a pixel requires a size of a square of 63.5 .mu.m. This size of pixel is obtained by convergence by a lens optical system not shown.
Even if the laser beam is converged to a desired size, the size of pixel may be enlarged if fixing is performed by a heat or pressure since toner is three-dimensionally applied on a photosensitive member in developing and fixing steps.
Thus, various technical problems exist as for one-pixel system and various proposals have been made as for the one-pixel system.
For example, Japanese Patent Application KOKAI Publication 58-34675 discloses that a laser beam needs to be shaped elliptically in a sub-scanning direction with respect to a desired size and one or two pixels in the sub-scanning direction are changed to correct this. Japanese Patent Application KOKAI Publication 58-107344 discloses a method of shortening a laser exposure period in the laser scanning direction to prevent pixels from being collapsed during fixing. Further, Japanese Patent Application KOKAI Publication 5-75816 discloses a method of making a correction by binarization based on power modulation from observation of two or more pixels prior to and subsequent to a target pixel.
Although various measures for correcting deterioration of image quality due to collapse of pixels have been introduced, all of the measures are one-dimensional correction measures and therefore cannot yet provide sufficient effects in response to deterioration of image quality due to collapse of characters.
Balance of an image in the longitudinal and lateral directions may be lost with correction in only the main scanning direction or sub-scanning direction.
Further, correction to pixels is advantageous for deterioration of image quality due to collapse of pixels in a character area, but leads to a problem of deterioration in tone of an image particularly in one-dimensional correction if the correction is made to a photographic image or an half-tone image since correction to pixels is effected on the entire image.
Although correction to a binary image advantageously requires a small circuit scale, advantageous measures have not been achieved to deal with multi-bit data.
As has been described above, various measures for correcting deterioration of image quality due to collapse of pixels during forming an image. However, all of the measures are one-dimensional and any of them cannot make sufficient effects on deterioration of image quality due to collapse of characters. With correction only in a main scanning direction or a sub-scanning direction, balance of an image in longitudinal and lateral directions may be broken.
Further, since the correction is effected on pixels of the entire image, advantages are obtained with respect to deterioration of image quality due to collapse of a character area. There is a problem that the tone of an image is deteriorated particularly in one-dimensional correction.
Another problem occurs in that correction to a binary image requires a circuit of a small scale but no effective measures are found to deal with multi-bit data.